import pandas as pd
import openpyxl
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.graph_objects as go

# Figure 6-A Pie chart (cell origins)

# EDA excel file is attached 
df = pd.read_excel(r"C:\Users\Mhaha\OneDrive\Desktop\EDA.xlsx", sheet_name='raw')

wdf_1 = df.iloc[1:, ].loc[:, ['Origin']] # wdf refers to working dataframe

# Extract the leading terms (e.g., "Animal", "Human", etc.) and Counting
wdf_1['Main_Term'] = wdf_1['Origin'].str.extract(r"^(\w+)")
wdf_1['Main_Term'].value_counts()

# the pie chart was made in the same Excel file (Pie charts sheet) using the data below

Main_Term
Animal     83
Human       8
Unclear     3
In          3
Both        1
Name: count, dtype: int64

# Figure 6-B Clustered vertical bars (Cell types and frequencies)

wdf_2 = df.iloc[1:,].iloc[:,[5]]

# Function to safely split and extract categories and details
def extract_category(cell_type):
    if isinstance(cell_type, str):
        return cell_type.split(' (')[0]
    return None  

def extract_details(cell_type):
    if isinstance(cell_type, str):
        # Extracting details inside parentheses, if any
        return pd.Series(cell_type).str.extract(r'\((.*?)\)').iloc[0, 0]
    return None

# Applying the functions to extract the category and details
wdf_2['Category'] = wdf_2['Cell_type'].apply(extract_category)
wdf_2['Details'] = wdf_2['Cell_type'].apply(extract_details)

# Pivoting the data to create columns for each category
pivoted_df = wdf_2.pivot(columns='Category', values='Details')

# Mergeing the new columns back with the original DataFrame
final_df = pd.DataFrame(pd.concat([wdf_2, pivoted_df], axis=1))
final_df

	Cell_type	Category	Details	NaN	Cell line	In vivo analysis	Other	Progenitor cells	Stem cells	Stem cells
1	Stem cells	Stem cells	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
2	Progenitor cells (NPCs)	Progenitor cells	NPCs	NaN	NaN	NaN	NaN	NPCs	NaN	NaN
3	Stem cells	Stem cells	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
4	Cell line	Cell line	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
5	Stem cells (NSCs)	Stem cells	NSCs	NaN	NaN	NaN	NaN	NaN	NSCs	NaN
...	...	...	...	...	...	...	...	...	...	...
95	Cell line	Cell line	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
96	Stem cells (MSCs)	Stem cells	MSCs	NaN	NaN	NaN	NaN	NaN	MSCs	NaN
97	Cell line	Cell line	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
98	Cell line	Cell line	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN
99	Cell line	Cell line	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN

99 rows × 10 columns

#Figure 6-B (continued)

#Rearrenging the columns
f_df = df.iloc[1:,].loc[:,['Cell_line']]
f_df_1 = final_df.iloc[:,6:9]
final_cell = pd.concat([f_df_1,f_df])

# Cleaning 
final_cell.columns = ['Other','Progenitor_cells','Stem_cells','Cell_line']
final_cell.Stem_cells = np.where(final_cell.Stem_cells ==  'BM-MSCs', 'MSCs',final_cell.Stem_cells )
final_cell.Stem_cells = np.where(final_cell.Stem_cells ==  'BMSCs', 'MSCs',final_cell.Stem_cells )
final_cell.Progenitor_cells = np.where(final_cell.Progenitor_cells ==  'IPSCs-derived neural spheroids', 'Neural spheroids',final_cell.Progenitor_cells )
final_cell.Other = np.where(final_cell.Other ==  'Organotypic\n(SC)', 'Organotypic (SC)',final_cell.Other )
final_cell.Other = np.where(final_cell.Other ==  'primary cells- Schwann cells', 'Schwann cells',final_cell.Other )

#Exporting to create visualization via R (refer to R code)
final_cell.to_csv('final_cell.csv')
final_cell

	Other	Progenitor_cells	Stem_cells	Cell_line
1	NaN	NaN	NaN	NaN
2	NaN	NPCs	NaN	NaN
3	NaN	NaN	NaN	NaN
4	NaN	NaN	NaN	NaN
5	NaN	NaN	NSCs	NaN
...	...	...	...	...
95	NaN	NaN	NaN	PC12
96	NaN	NaN	NaN	NaN
97	NaN	NaN	NaN	SH-SY5Y
98	NaN	NaN	NaN	SH-SY5Y
99	NaN	NaN	NaN	NaN

198 rows × 4 columns

# Figure 6-C (Distribution of Studies by Exact Cell density and Unit)

# cleaning cell density unit data 
wdf_3 = df.iloc[1:,].loc[:,['Cell_density_unit']]
wdf_3 = wdf_3.dropna().reset_index().drop(columns = 'index')
wdf_3.Cell_density_unit =wdf_3.Cell_density_unit.str.title()
wdf_3.Cell_density_unit= np.where(wdf_3.Cell_density_unit ==  'Cell/Ml', 'Cells/Ml',wdf_3.Cell_density_unit )
wdf_3.Cell_density_unit= np.where(wdf_3.Cell_density_unit ==  'Ml', 'Cells/Ml',wdf_3.Cell_density_unit )
wdf_3.Cell_density_unit= np.where(wdf_3.Cell_density_unit ==  'Cell/Well', 'Cells/Well',wdf_3.Cell_density_unit )

# cleaning cell density value data
wdf_4 =df.iloc[1:,].loc[:,['Cell_denisty']]
wdf_4.Cell_denisty = wdf_4.Cell_denisty.astype(float)
wdf_4 = wdf_4.dropna().reset_index().drop(columns='index')

# Merging the two data
density_df = pd.concat([wdf_3,wdf_4],axis=1)

# Formatting cell density values to scientific notation

def format_density_density(value):  
    """Convert a numerical value to the format 's x 10^n'."""  
    if value == 0:  
        return "0"  
    else:  
        exponent = int(np.floor(np.log10(value)))   
        coefficient = value / (10 ** exponent)       
        if coefficient >= 10:  
            coefficient /= 10  
            exponent += 1  
        return f"{coefficient:.1f} x 10^{exponent}"  

# Apply the function to create a new column 'Formatted_Density'  
density_df['Formatted_Density'] = density_df['Cell_denisty'].apply(format_density_density)  

# creating the final df 
density_df = pd.DataFrame(density_df[['Cell_density_unit', 'Cell_denisty', 'Formatted_Density']])
density_df = density_df[density_df.Cell_density_unit != 'Cells/Disc']
density_df = density_df[density_df.Cell_density_unit != 'Cells/Cm2']
density_df

	Cell_density_unit	Cell_denisty	Formatted_Density
0	Cells/Well	1000000.0	1.0 x 10^6
1	Cells/Well	10000000.0	1.0 x 10^7
3	Cells/Well	10000.0	1.0 x 10^4
4	Cells/Hydrogel	10000.0	1.0 x 10^4
5	Cells/Ml	1000000.0	1.0 x 10^6
...	...	...	...
64	Cells/Hydrogel	200000.0	2.0 x 10^5
65	Cells/Ml	10000000.0	1.0 x 10^7
66	Cells/Hydrogel	200000.0	2.0 x 10^5
67	Cells/Hydrogel	5000000.0	5.0 x 10^6
68	Cells/Ml	10000.0	1.0 x 10^4

64 rows × 3 columns

# Figure 6-C (Continued: Graphing)
grouped = density_df.groupby(['Formatted_Density', 'Cell_density_unit']).size().unstack(fill_value=0)  

# Extracting numeric value for sorting y-axis
def extract_numeric_value(formatted_value):  
    """Extracts a numerical value from the formatted density string."""  
    if formatted_value == "0":  
        return 0  
    coefficient, exponent = formatted_value.split(' x 10^')  
    return float(coefficient) * (10 ** int(exponent))  

# applying the function to sort the values 
grouped['Numeric_Value'] = grouped.index.map(extract_numeric_value)  
grouped = grouped.sort_values(by='Numeric_Value')  
grouped = grouped.drop(columns='Numeric_Value')  

# plotting horizontal stacked bar chart with specified dimensions  
colors = ['lightgrey', 'lightblue', 'lightgreen']  
ax = grouped.plot(kind='barh', stacked=True, color=colors, edgecolor='black', figsize=(14, 5), width=0.7)  # Set figure size here  

# Adding labels,title, and legend
plt.xlabel('Number of Studies', fontsize=17)  
plt.ylabel('Cell Density', fontsize=17,labelpad=10)  
plt.title('Distribution of Studies by Exact Cell Density and Unit', fontsize=16, pad=20,fontweight='bold')  
plt.yticks(fontsize=11, va='center', rotation=0, ha='right')
plt.xticks(size=10)
plt.subplots_adjust(left=0.2)  
plt.legend(title='Cell Density Unit', loc='upper right', fontsize=12, bbox_to_anchor=(0.5, 0.05, 0.5, 0.5))
plt.grid(axis='x', linestyle='--', linewidth=0.7)  

# adjusting the y-tick labels position
for label in ax.get_yticklabels():  
    label.set_x(label.get_position()[0] - 0.007)  
    
# Annotating counts inside the bars 
for i in range(grouped.shape[0]):  
    cumulative_sum = grouped.iloc[i].cumsum()  
    counts = grouped.iloc[i]  
    for j in range(len(counts)):  
        if counts[j] > 0:    
            ax.text(cumulative_sum[j] - counts[j] * 0.03, i, str(counts[j]),  
                    ha='right', va='center', fontsize=9, color='black')  
            
#plt.savefig('filename1.png',bbox_inches='tight')  # Saving the file 
plt.tight_layout()
plt.show()

C:\Users\Mhaha\AppData\Local\Temp\ipykernel_26048\314692049.py:40: FutureWarning:

Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

C:\Users\Mhaha\AppData\Local\Temp\ipykernel_26048\314692049.py:41: FutureWarning:

Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

# Figure 6-D (Cell culture Mode, Duration, and seeding Unit)

# Cleaning the data for mode and duration of culture
wdf_5 = df.iloc[1:,].loc[:,['Mode','Culture_duration']]
wdf_5 =wdf_5.dropna().reset_index().drop(columns='index')
wdf_5 = wdf_5[wdf_5.Culture_duration !='Not reported']

# merging with the cell density unit data
wdf_6 = pd.concat([wdf_5,wdf_3],axis=1) 

#Exporting to create visualization via R (refer to R code)
wdf_6.to_csv('wdf_6.csv')

# Annotations were added using the counts obtained below
print(wdf_6.Mode.value_counts())
print('______________')
print(wdf_6.Culture_duration.value_counts())
print('______________')
print(wdf_6.Cell_density_unit.value_counts())

wdf_6

Mode
Mono-culture    87
Co-culture       4
Name: count, dtype: int64
______________
Culture_duration
3d-2w    76
2w-3m    10
<3d       5
Name: count, dtype: int64
______________
Cell_density_unit
Cells/Ml          24
Cells/Hydrogel    23
Cells/Well        17
Cells/Cm2          4
Cells/Disc         1
Name: count, dtype: int64

	Mode	Culture_duration	Cell_density_unit
0	Mono-culture	3d-2w	Cells/Well
1	Mono-culture	2w-3m	Cells/Well
2	Co-culture	3d-2w	Cells/Disc
3	Mono-culture	3d-2w	Cells/Well
4	Mono-culture	2w-3m	Cells/Hydrogel
...	...	...	...
88	Mono-culture	3d-2w	NaN
89	Mono-culture	3d-2w	NaN
90	Mono-culture	2w-3m	NaN
91	Mono-culture	2w-3m	NaN
33	NaN	NaN	Cells/Hydrogel

92 rows × 3 columns

# Fig 7-A Pie chart (Counts below ere used to make the core/ refer to Pie charts sheet in EDA)
wdf_7 = df.iloc[1:,].loc[:,['Hydrogel_origion']]
print(wdf_7.dropna().value_counts())
print('___________________')

wdf_8 = df.iloc[1:,].loc[:,['Natural_type']]
wdf_8 = wdf_8.dropna()
wdf_8.Natural_type = np.where(wdf_8.Natural_type =='protein-based','Protein-based' ,wdf_8.Natural_type)
wdf_8.Natural_type.value_counts()

Hydrogel_origion         
Natural                      76
Synthetic                    12
Mixed (Natural,Synthetic)     9
Mixed                         1
Name: count, dtype: int64
___________________

Natural_type
Polysaccharide-based                           52
Protein-based                                  12
Mixed(Protein-based, Polysaccharide-based)     10
Mixed(ECM-based, Polysaccharide-based)          1
Mixed                                           1
Name: count, dtype: int64

# Fig 7-A Pie chart (percentages below are used to make the periphery of Polysccharide-based/ refer to Pie charts sheet in EDA)

# Subseting the data containing only Polysaccharides
wdf_9 = df.iloc[1:,].loc[:,['Natural_type','Polysaccharide_type','Mixed_hydrogel_type']]
wdf_9 = wdf_9[wdf_9['Natural_type'] =='Polysaccharide-based']

# Separating multiple Polysaccharides reported with a "comma" between them and stacking all the indiviual types
wdf_9_sep= wdf_9['Mixed_hydrogel_type'].str.split(',', expand=True)
wdf_9_stacked = pd.DataFrame(pd.concat([wdf_9.iloc[:,1],wdf_9_sep.loc[:,0],wdf_9_sep.loc[:,1]], axis=0))
wdf_9_stacked = wdf_9_stacked.reset_index().rename({0:'Polysac_name'}, axis=1).drop(columns='index').dropna()

# Cleaning and Calculating the count of occurance
wdf_9_stacked.Polysac_name = wdf_9_stacked.Polysac_name.str.strip().str.title() 
wdf_9_stacked.value_counts() 

Polysac_name    
Gelatin             28
Chitosan            12
Alginate            10
Ha                   7
Cellulose            2
Agarose              1
Chitin               1
Gellan Gum           1
Other (Pullulan)     1
Xanthan Gum          1
Name: count, dtype: int64

# Fig 7-A Pie chart (percentages below are used to make the periphery of Protein-based/ refer to Pie charts sheet in EDA)

# Subseting the data contianing only Proteins
wdf_10 = df.iloc[1:,].loc[:,['Natural_type','Protein_type','Mixed_hydrogel_type']]
wdf_10 = wdf_10[wdf_10['Natural_type'] =='Protein-based']

# Separating multiple Proteins reported with a "comma" between them and stacking all the indiviual types
wdf_10_sep= wdf_10['Mixed_hydrogel_type'].str.split(',', expand=True)
wdf_10_stacked = pd.DataFrame(pd.concat([wdf_10.iloc[:,1],wdf_10_sep.loc[:,0],wdf_10_sep.loc[:,1]], axis=0))
wdf_10_stacked = wdf_10_stacked.reset_index().rename({0:'Protein_name'}, axis=1).drop(columns='index').dropna()

# Cleaning and Calculating the count of occurance
wdf_10_stacked.Protein_name = wdf_10_stacked.Protein_name.str.strip().str.title() 
wdf_10_stacked.value_counts() 

Protein_name
Silk            7
Collagen        5
Fibrin          1
Name: count, dtype: int64

# Fig 7-A Pie chart (percentages below are used to make the periphery of Mixed/ refer to Pie charts sheet in EDA)
wdf_11 = df.iloc[1:,].loc[:,['Mixed_hydrogel_type']]

# Cleaning and sorting alphabetically
wdf_11.Mixed_hydrogel_type = wdf_11.Mixed_hydrogel_type.str.replace(r'\s+', '', regex=True).str.title()
def sort_comma_separated_values(value):
    if pd.isna(value):  # Check if the value is NaN
        return value
    return ','.join(sorted(value.split(',')))

# Applying the function and calculating count of occurance 
wdf_11['Sorted_Mixed_hydrogel_type'] = wdf_11['Mixed_hydrogel_type'].apply(sort_comma_separated_values)
wdf_11.Sorted_Mixed_hydrogel_type.value_counts() 

Sorted_Mixed_hydrogel_type
Chitosan,Gelatin           5
Collagen,Ha                5
Alginate,Gelatin           3
Collagen,Gelatin           2
Chitosan,Polyurethane      2
Gelatin,Ha                 2
Chitosan,Silk              1
Agarose,Gelatin            1
Fibrinogen,Peo             1
Chitosan,Ha                1
Chitosan,Paam              1
Ha,Paam                    1
Matrigel                   1
Ecm(Spinalcord),Gelatin    1
Gelatin,Gellangum,Peg      1
Alginate,Collagen          1
Peg,Silk                   1
Aam,Gelma,Nipam            1
Cellulose,Chitosan         1
Alginate,Cellulose         1
Collagen,Silk              1
Peg,Polyurethane           1
Alginate,Ha,Poloxamer      1
Gelatin,Silk               1
Pcl,Silk                   1
Name: count, dtype: int64

# Fig 7-B1 (Nanomaterial type and Frequencies)

# Data is manually cleaned in 'Clean' sheet of EDA excel file
df_1 = pd.read_excel(r"C:\Users\Mhaha\OneDrive\Desktop\EDA.xlsx", sheet_name='Clean')
wdf_12 = df_1.iloc[1:,].loc[:,['Carbon_type', 'Metal_type', 'Ceramic_type', 'Polymer_type', 'Semiconductor_type']]

# More cleaning
wdf_12.Polymer_type= np.where(wdf_12.Polymer_type == 'Polythiophene', 'PTh',wdf_12.Polymer_type)
wdf_12.Ceramic_type= np.where(wdf_12.Ceramic_type== 'Bismuth errite', 'BiFeO3',wdf_12.Ceramic_type)
wdf_12.Ceramic_type= np.where(wdf_12.Ceramic_type== 'Cobalt ferrite', 'CoFe2O4',wdf_12.Ceramic_type)
wdf_12.Ceramic_type= np.where(wdf_12.Ceramic_type== 'Barium Titanate', 'BaTiO3',wdf_12.Ceramic_type)

# Exporting for graphing in R (refer to R code)
wdf_12.to_csv("wdf_12.csv")
wdf_12

	Carbon_type	Metal_type	Ceramic_type	Polymer_type	Semiconductor_type
1	NaN	NaN	NaN	NaN	BP
2	CNT	NaN	NaN	NaN	NaN
3	CNT	NaN	NaN	NaN	NaN
4	Graphene	NaN	BaTiO3	NaN	NaN
5	CNT	NaN	NaN	NaN	NaN
...	...	...	...	...	...
96	CNT	NaN	NaN	NaN	NaN
97	GO	NaN	BiFeO3	NaN	NaN
98	NaN	NaN	CoFe2O4	NaN	NaN
99	NaN	NaN	NaN	NaN	NaN
100	NaN	NaN	NaN	NaN	NaN

100 rows × 5 columns

# Fig 7-B2 (Nanomaterial Type Distribution)

#Data is manually cleaned in 'Clean' sheet of EDA excel file
df_1 = pd.read_excel(r"C:\Users\Mhaha\OneDrive\Desktop\EDA.xlsx", sheet_name='Clean')
wdf_13 = df_1.iloc[1:,].loc[:,['Nanomaterial_type']]
wdf_13 = wdf_13.dropna(subset='Nanomaterial_type')
wdf_13.Nanomaterial_type =wdf_13.Nanomaterial_type.str.title().str.strip()

# More cleaning
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type == 'Mixed (Ceramic-Based,Metal-Based)','Mixed (Metal-Based, Ceramic-Based)',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type == 'Mixed(Carbon-Based, Semiconductor-Based)', 'Mixed (Carbon-Based, Semiconductor-Based)',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type =='Mixed ','Mixed',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type == 'Carbon-Based','Carbon',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type == 'Ceramic-Based','Ceramic',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type ==  'Polymer-Based','Polymer',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type ==  'Metal-Based','Metal',wdf_13.Nanomaterial_type)
wdf_13.Nanomaterial_type = np.where(wdf_13.Nanomaterial_type ==  'Semiconductor-Based','Semiconductor',wdf_13.Nanomaterial_type)

# Exporting for graphing in R (refer to R code)
wdf_13.to_csv("wdf_13.csv")
wdf_13

	Nanomaterial_type
1	Semiconductor
2	Carbon
3	Carbon
4	Ceramic
5	Carbon
...	...
94	Metal
95	Mixed
96	Carbon
97	Carbon
98	Metal

98 rows × 1 columns

# Fig 7-C1 (Snakey graph at top: cleaning)
wdf = df.iloc[1:,].loc[:,['Hydrogel_origion',"Natural_type","Crosslinking_type"]].dropna(subset="Crosslinking_type")

# cleaning the Crosslinking_type column 
wdf.Crosslinking_type = wdf.Crosslinking_type.str.title()
wdf.Crosslinking_type = np.where(wdf.Crosslinking_type == 'covalent', 'Covalent',wdf.Crosslinking_type)
wdf.Crosslinking_type = np.where(wdf.Crosslinking_type == 'Physical ', 'Physical',wdf.Crosslinking_type)
wdf = wdf[wdf.Crosslinking_type.isin(['Covalent','Ionic','Photopolymerization','3D Bioprinting','Physical','Self-Assembly'])]

# cleaning the Natural_type column 
wdf.Natural_type = wdf.Natural_type.str.title()
df = wdf[~wdf.Natural_type.isin(['Mixed(Protein-Based, Polysaccharide-Based) ', 'Mixed(Ecm-Based, Polysaccharide-Based) ','Mixed' ])]

# cleaning the Hydrogel_origion column 
wdf.Hydrogel_origion = wdf.Hydrogel_origion.str.title()
wdf = wdf[~wdf.Natural_type.isin(['Natural', 'Mixed', ])]
wdf.Hydrogel_origion = np.where(wdf.Hydrogel_origion =='Natural', np.nan, wdf.Hydrogel_origion)
wdf = wdf.reset_index().drop(columns='index')
wdf.sample(10)

# merging the first two columns
wdf['Merged_Column'] = wdf['Hydrogel_origion'].fillna(wdf['Natural_type'])
wdf = wdf.loc[:,["Merged_Column", 'Crosslinking_type']]
wdf.sample(10)

	Merged_Column	Crosslinking_type
54	Polysaccharide-Based	Covalent
18	Polysaccharide-Based	Covalent
45	Mixed (Natural,Synthetic)	Covalent
12	Synthetic	Photopolymerization
63	Mixed(Protein-Based, Polysaccharide-Based)	Physical
51	Mixed (Natural,Synthetic)	Ionic
17	Polysaccharide-Based	Covalent
4	Polysaccharide-Based	Photopolymerization
39	Mixed(Ecm-Based, Polysaccharide-Based)	Photopolymerization
1	Mixed	Photopolymerization

# Prepareing the data
source_labels = wdf.iloc[:, 0].unique().tolist() 
target_labels = wdf.iloc[:, 1].unique().tolist()  
all_labels = source_labels + target_labels

# Creating a mapping from label to index
label_to_index = {label: i for i, label in enumerate(all_labels)}
wdf.loc[:, 'source_index'] = wdf.iloc[:, 0].map(label_to_index)
wdf.loc[:, 'target_index'] = wdf.iloc[:, 1].map(label_to_index)

# Creating the Sankey diagram 
fig = go.Figure(data=[go.Sankey(
    node=dict(
        pad=15,
        thickness=20,
        line=dict(color="black", width=0.5),
        label=all_labels
    ),
    link=dict(
        source=wdf['source_index'],  
        target=wdf['target_index'],  
        value=[1] * len(wdf)  
    )
)])


fig.update_layout(

    font=dict(
        size=21,
        color="black",
        family="Arial"  
    ),
    width=600,
    height=600,
    margin=dict(l=10, r=10, t=10, b=10)
)


fig.write_image("sankey_diagram.png", width=500 ,height=500)

fig.show()

# Fig 7-C2 (Snakey graph at bottom: cleaning)
wdf_14 =df.iloc[1:,].loc[0:,['Hydrogel_geometry','Tissue_modeled']]
wdf_14 = wdf_14.dropna()
wdf_14.Hydrogel_geometry = wdf_14.Hydrogel_geometry.str.title()
wdf_14 = wdf_14[wdf_14.Hydrogel_geometry.isin(['General 3D Structure','Conduit','Injectable','Film/Sheet'])]

# Cleaning
wdf_14.Tissue_modeled = wdf_14.Tissue_modeled.str.upper()
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'CNS (SCI)', 'CNS(SCI)',wdf_14.Tissue_modeled)
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'CNS (SCI)', 'CNS(SCI)',wdf_14.Tissue_modeled)
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'CSN (SCI)', 'CNS (SCI)',wdf_14.Tissue_modeled)
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'CSN (SCI)', 'CNS (SCI)',wdf_14.Tissue_modeled)
wdf_14 = wdf_14[wdf_14.Tissue_modeled !=  '·\xa0\xa0\xa0\xa0\xa0\xa0\xa0 UNCLEAR']
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'CNS(SCI)', 'CNS (SCI)',wdf_14.Tissue_modeled)
wdf_14 = wdf_14[wdf_14.Tissue_modeled != 'CNS']
wdf_14 = wdf_14[wdf_14.Tissue_modeled != 'PNS ( NEUROGENESIS IN MUSCULATURE)']
wdf_14 = wdf_14[wdf_14.Tissue_modeled != 'PNS']
wdf_14 = wdf_14[wdf_14.Tissue_modeled != 'OTHER(BIOINTERFACE)']
wdf_14 = wdf_14[wdf_14.Tissue_modeled != ' UNCLEAR']
wdf_14.Tissue_modeled = np.where(wdf_14.Tissue_modeled == 'OTHER (GENERAL DIFFERENTIATION)', 'GENERAL DIFFERENTIATION',wdf_14.Tissue_modeled)
wdf_14

	Hydrogel_geometry	Tissue_modeled
1	General 3D Structure	CNS (SCI)
3	Conduit	GENERAL DIFFERENTIATION
4	Conduit	PNS (SCIATIC NERVE)
5	General 3D Structure	GENERAL DIFFERENTIATION
6	General 3D Structure	CNS (SCI)
...	...	...
94	Conduit	PNS (SCIATIC NERVE)
95	General 3D Structure	CNS (SCI)
96	Injectable	CNS (BRAIN)
97	General 3D Structure	GENERAL DIFFERENTIATION
98	General 3D Structure	CNS (BRAIN)

82 rows × 2 columns

# Fig 7-C2 (Snakey graph at bottom: graphing)


target_labels = wdf_14['Hydrogel_geometry'].unique().tolist()  
source_labels = wdf_14['Tissue_modeled'].unique().tolist()  

# Create a combined list of all unique labels
all_labels = source_labels + target_labels

# Create a mapping from label to index
label_to_index = {label: i for i, label in enumerate(all_labels)}

# Map source and target columns to index values
wdf_14['source_index'] = wdf_14['Tissue_modeled'].map(label_to_index)  
wdf_14['target_index'] = wdf_14['Hydrogel_geometry'].map(label_to_index) 

link_counts = wdf_14.groupby(['source_index', 'target_index']).size().reset_index(name='count')

# Create the Sankey diagram
fig = go.Figure(data=[go.Sankey(
    node=dict(
        pad=15,
        thickness=20,
        line=dict(color="black", width=0.5),
        label=all_labels,  
    ),
    link=dict(
        source=link_counts['source_index'],  # Indices of the source nodes
        target=link_counts['target_index'],  # Indices of the target nodes
        value=link_counts['count']            # Use the counts for the connections
    )
)])

fig.update_layout(

    font=dict(
        size=17,
        color="black",
        family="Arial"  
    ),
    width=600,
    height=600,
    margin=dict(l=10, r=10, t=10, b=10)
)

fig.write_image("sankey_diagram_flipped.png", width=500, height=500, scale=2)  # Higher resolution

fig.show()

# Fig 7-D (Nanomaterials' Integration and Dimension)

# Subseting Column for nanotype dimensions
wdf_15_a = df_1.iloc[1:,].loc[:,['Nanomat_dimension']]

# Separate by commas and expand into separate rows
df_expanded = wdf_15_a['Nanomat_dimension'].str.split(',', expand=True).stack().reset_index(drop=True)
df_combined = pd.DataFrame(df_expanded, columns=['Nanomat_dimension'])
df_combined['Nanomat_dimension'] = df_combined['Nanomat_dimension'].str.strip()
df_combined = df_combined[df_combined.Nanomat_dimension != 'Unclear']

# Subseting Column for nanotype integration methods
wdf_15_b =df_1.iloc[1:,].loc[:,['Nano_Hyd_integration']]
wdf_15_b = wdf_15_b .reset_index().drop(columns='index').dropna()

#combining and cleaning final dataframe
wdf_15 = pd.concat([wdf_15_b,df_combined], axis=1)
wdf_15.Nano_Hyd_integration = np.where(wdf_15.Nano_Hyd_integration == 'Other ( placement after hydrogel synthesis)','Other',wdf_15.Nano_Hyd_integration)
wdf_15.Nano_Hyd_integration = np.where(wdf_15.Nano_Hyd_integration == 'Other (layered)','Other',wdf_15.Nano_Hyd_integration)
wdf_15.Nano_Hyd_integration = np.where(wdf_15.Nano_Hyd_integration == 'Other ','Other',wdf_15.Nano_Hyd_integration)

# Exporting for graphing in R (refer to R code)
wdf_15.to_csv("wdf_15.csv")
wdf_15

# The counts below were used for annotaion
#wdf_15.Nanomat_dimension.value_counts()
#wdf_15.Nano_Hyd_integration.value_counts()

	Nano_Hyd_integration	Nanomat_dimension
0	OP1	2D
1	OP1	1D
2	OP1	2D
3	Other	1D
4	OP1	0D
...	...	...
103	NaN	0D
104	NaN	0D
105	NaN	1D
106	NaN	2D
107	NaN	0D

108 rows × 2 columns

# Fig 8-A1 (In Vitro Hydrogel Characterization Methods)

wdf_16 = df.iloc[1:,].loc[:,['In_vitro_hyd_assay']]

# Separate by commas and expand into separate rows
wdf_16_a = wdf_16['In_vitro_hyd_assay'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_16_b = pd.DataFrame(wdf_16_a, columns=['In_vitro_hyd_assay'])
wdf_16_b['In_vitro_hyd_assay'] = wdf_16_b['In_vitro_hyd_assay'].str.strip().str.title()

# Display the resulting DataFrame
wdf_16_b.In_vitro_hyd_assay.value_counts()

In_vitro_hyd_assay
Mechanical Test        62
Electrical Property    57
Rheological Test       49
Swelling Test          44
Degredation Test       31
Porosity               26
Biocompatibility        1
                        1
Name: count, dtype: int64

# Fig 8-A2 (In Vitro Cell Characterization Methods)
wdf_17 = df.iloc[1:,].loc[:,['In_vitro_cell_assay']]


# Split the 'In_vivo_asess_method' column by commas and expand into separate rows
wdf_17_a = wdf_17['In_vitro_cell_assay'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_17_b = pd.DataFrame(wdf_17_a, columns=['In_vitro_cell_assay'])
wdf_17_b['In_vitro_cell_assay'] = wdf_17_b['In_vitro_cell_assay'].str.strip().str.lower()

# Display the resulting DataFrame
wdf_17_b

	In_vitro_cell_assay
0	cell viability (l/d staining)
1	cell proliferation (cck8)
2	neural protein expression analysis
3	cell viability(l/d staining)
4	neural protein expression analysis
...	...
289	cell migraiton
290	neural protein expression analysis
291	neural gene expression analysis
292	cell viability(l/d staining - mtt)
293	neural protein expression analysis

294 rows × 1 columns

# Fig 8-A2 (Continued)
cell_viability = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('cell viability', na=False)]
cell_proliferation = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('cell proliferation', na=False)]
protein_expression = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('protein expression', na=False)]
gene_expression = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('gene expression', na=False)]
cell_adhesion = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('cell adhesion', na=False)]
cell_migration = wdf_17_b[wdf_17_b['In_vitro_cell_assay'].str.contains('cell migration', na=False)]

# Using str.contains and value.counts the freuency of specific methods used in each category were obtained
in_vit_cell = pd.DataFrame({
    "Neural Protein Expression Analysis_Total" : [87],
    "In Vitro (Neural Protein Expression Analysis)": [73],
    "In Vivo (Neural Protein Expression Analysis)" : [14],
    "Neural Gene Expression Analysis" : [37],
    "Cell Viability: Total" : [90],
    "Cell Viability: L/D staining": [52],
    "Cell Viability: MTT": [18],
    "Cell Viability: Other": [20],
    "Cell Proliferation_Total": [53],
    "Cell Proliferation: CCK-8": [40],
    "Cell Proliferation: Other": [13],
    "Cell Adhesion": [21],
    "Cell Migration_Total": [8],
    "Cell Migration: Transwell": [5],
    "Cell Migration: Other": [3]
})

#The dataframe above was then defined in R for graphing

# The look of the dataframes obtained after filtering
pd.concat([cell_viability, cell_proliferation,protein_expression, cell_adhesion,cell_migration ], axis=1).sample(10)

	In_vitro_cell_assay	In_vitro_cell_assay	In_vitro_cell_assay	In_vitro_cell_assay	In_vitro_cell_assay
84	NaN	NaN	neural protein expression analysis	NaN	NaN
183	NaN	NaN	neural protein expression analysis	NaN	NaN
38	NaN	NaN	neural protein expression analysis	NaN	NaN
164	NaN	NaN	NaN	cell adhesion	NaN
288	NaN	NaN	NaN	cell adhesion	NaN
144	NaN	NaN	neural protein expression analysis	NaN	NaN
142	cell viability( l/d staining)	NaN	NaN	NaN	NaN
186	NaN	NaN	NaN	NaN	cell migration ( transwell)
211	NaN	cell proliferation (cck8)	NaN	NaN	NaN
52	cell viability (l/d staining)	NaN	NaN	NaN	NaN

# Fig 8-B (Pie chart)

wdf_18 = df.iloc[1:,].loc[:,['mech_parameter_type']]

# Splitting the values in column by commas and expanding into separate rows
wdf_18_a = wdf_18['mech_parameter_type'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_18_b = pd.DataFrame(wdf_18_a, columns=['mech_parameter_type'])
wdf_18_b['mech_parameter_type'] = wdf_18_b['mech_parameter_type'].str.strip().str.lower()

wdf_18_b['mech_parameter_type'].value_counts() 

# the pie chart is made using the percentages below ( refer to Pie charts Sheet)

mech_parameter_type
storage or loss modulus (g’/g’’)    42
young modulus                       31
compressive modulus                 22
tensile strength                     6
tensile strain                       1
other( steady-state moduli)          1
Name: count, dtype: int64

# Fig 8-C 
wdf_19 = df.iloc[1:,].loc[:,['Elec_method']]

# Splitting the 'Elec_method' column by commas and expanding into separate rows
wdf_19_a = wdf_19['Elec_method'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_19_b = pd.DataFrame(wdf_19_a, columns=['Elec_method'])
wdf_19_b['Elec_method'] = wdf_19_b['Elec_method'].str.strip().str.lower()

conductivity = wdf_19_b[wdf_19_b['Elec_method'].str.contains('conductivity', na=False)]
impedance = wdf_19_b[wdf_19_b['Elec_method'].str.contains('impedance', na=False)]
electrochemical = wdf_19_b[wdf_19_b['Elec_method'].str.contains('electrochemical', na=False)]

# Using str.contains and value.counts the percent freuency of specific methods used in each category were obtained

print('For conductivity')
print(len(conductivity) / len(wdf_19.dropna()))
print(len(conductivity[conductivity['Elec_method'].str.contains('4 probe', na=False)]) / len(conductivity))
print(len(conductivity[conductivity['Elec_method'].str.contains('2 probe', na=False)]) / len(conductivity))
print('_____________________')
print('For Impedance')
print(len(impedance) / len(wdf_19.dropna()))
print(len(impedance[impedance['Elec_method'].str.contains('eis', na=False)]) / len(impedance))
print('_____________________')
print('For electrochemical')
print(len(electrochemical) / len(wdf_19.dropna()))
print(len(electrochemical[electrochemical['Elec_method'].str.contains('cv', na=False)]) / len(electrochemical))

For conductivity
0.8035714285714286
0.6222222222222222
0.15555555555555556
_____________________
For Impedance
0.3392857142857143
0.3684210526315789
_____________________
For electrochemical
0.05357142857142857
0.6666666666666666

# Fig 8-D (data cleaning)

wdf_20 = df.iloc[1:,].loc[:,['Tissue_modeled','Conductivity']]

# Cleaning before graphing
wdf_20_a = wdf_20.dropna(subset='Conductivity').drop([3]).reset_index().drop(columns='index')
wdf_20_a.Tissue_modeled = np.where(wdf_20_a.Tissue_modeled =='CNS(SCI)','CNS (SCI)',wdf_20_a.Tissue_modeled )
wdf_20_a.Tissue_modeled = np.where(wdf_20_a.Tissue_modeled =='Other (general differentiation)','General Differentiation',wdf_20_a.Tissue_modeled )
wdf_20_a= wdf_20_a[wdf_20_a.Tissue_modeled.isin(['CNS (SCI)' ,'General Differentiation','PNS (Sciatic nerve)','CNS (Brain)'])]

# rearranging and dropping extreme outliers
wdf_20_b = wdf_20_a.pivot_table(index= wdf_20_a.index, columns='Tissue_modeled', values ='Conductivity')
wdf_20_b.drop([12,31], inplace= True)
wdf_20_b.head(10)

Tissue_modeled	CNS (Brain)	CNS (SCI)	General Differentiation	PNS (Sciatic nerve)
0	NaN	0.027	NaN	NaN
2	NaN	0.0021	NaN	NaN
3	NaN	NaN	0.31	NaN
4	NaN	NaN	0.000087	NaN
5	NaN	0.004	NaN	NaN
7	NaN	0.004	NaN	NaN
8	NaN	0.000013	NaN	NaN
9	NaN	NaN	0.0033	NaN
10	NaN	NaN	0.00152	NaN
13	NaN	NaN	2.132	NaN

# Fig 8-D (Continued : graphing)

# Melting the DataFrame to convert it from wide to long format
melted_df = wdf_20_b.reset_index(drop=True).melt(var_name='Tissue Model', value_name='Value')

# Creating the boxplot 
plt.figure(figsize=(8, 4))  
sns.boxplot(x='Value', y='Tissue Model', data=melted_df, width=0.8)  
plt.xscale('log')

# Adding labels and title with padding
plt.xlabel('Log(Conductivity) (σ)', fontsize=20)  
plt.ylabel('', fontsize=20)  
plt.tick_params(axis='both', labelsize=18)  
plt.subplots_adjust(left=0.15)
plt.tight_layout()
#plt.savefig('Con.png')

plt.show()

# Fig 9-A

wdf_21 = df.iloc[1:,].loc[:,['diff_condition']]


# Splitting the values in the column by commas and expanding into separate rows
wdf_21_a = wdf_21['diff_condition'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_21_b = pd.DataFrame(wdf_21_a, columns=['diff_condition'])
wdf_21_b['diff_condition'] = wdf_21_b['diff_condition'].str.replace(r'\s+', '', regex=True).str.upper()

# The Venn diagram was made using the counts below
wdf_21_b.diff_condition.value_counts()

diff_condition
C            38
C+E          26
C+N          26
C+N+E         9
INVIVO(C)     1
C+N/C+N+E     1
Name: count, dtype: int64

# Fig 9-B (upper graph)


# Splitting the values in the column by commas and expanding into separate rows
wdf_22 = df.iloc[1:,].loc[:,['Neurotrophic_factor']]
wdf_22_a = wdf_22['Neurotrophic_factor'].str.replace('Mixed (', '', regex=False).str.split(',', expand=True).stack().reset_index(drop=True)
wdf_22_b = pd.DataFrame(wdf_22_a, columns=['Neurotrophic_factor'])
wdf_22_b['Neurotrophic_factor'] = wdf_22_b['Neurotrophic_factor'].str.strip().str.upper()

# Calculating the vounts
print(wdf_22_b[wdf_22_b.Neurotrophic_factor.str.contains('NGF')].value_counts()/ len(wdf_22.dropna()))
print('_____________')
print(wdf_22_b[wdf_22_b.Neurotrophic_factor.str.contains('RA')].value_counts()/ len(wdf_22.dropna()))
print('_____________')
print(wdf_22_b[wdf_22_b.Neurotrophic_factor.str.contains('BFGF')].value_counts()/ len(wdf_22.dropna()))
print('_____________')
print(wdf_22_b[wdf_22_b.Neurotrophic_factor.str.contains('BDNF')].value_counts()/ len(wdf_22.dropna()))

Neurotrophic_factor
NGF                    0.314286
Name: count, dtype: float64
_____________
Neurotrophic_factor
RA                     0.200000
RETINOIC ACID (RA)     0.057143
Name: count, dtype: float64
_____________
Neurotrophic_factor                  
BFGF                                     0.142857
BASIC FIBROBLAST GROWTH FACTOR (BFGF)    0.028571
Name: count, dtype: float64
_____________
Neurotrophic_factor
BDNF                   0.085714
Name: count, dtype: float64

# Fig 9-B (lower graph)
wdf_23 = df.iloc[1:,].loc[:,['Elec_stimulation_type']]

# The percentages below were used to create the graph in R 
wdf_23.Elec_stimulation_type.value_counts() / len(wdf_23.dropna()) * 100

Elec_stimulation_type
Wireless electrical stimulation (magnetic)      36.111111
DC                                              33.333333
AC                                              11.111111
Wireless electrical stimulation (optical)        8.333333
Wireless electrical stimulation (ultrasound)     5.555556
DC (pulsed)                                      2.777778
AC (Biphasic)                                    2.777778
Name: count, dtype: float64

# Fig 9-C1 (data cleaning)
wdf_24 = df_1.iloc[1:,].loc[:,['Neurotrophic_factor','NF_days']].dropna(subset='NF_days')

# Dropping 'Not reported' values
wdf_24 = wdf_24.drop([27,38,40,59],axis =0)

#rearranging data for graphing
wdf_24_a = wdf_24.pivot_table(index=wdf_24.index, columns='Neurotrophic_factor', values='NF_days', aggfunc='first')
wdf_24_a['All'] = wdf_24_a.apply(lambda row: row.sum(), axis=1)
wdf_24_a

Neurotrophic_factor	BDNF	BDNF, GDNF	NGF	NeuroCult	Neurobasal A medium	Not reported	RA	VEGF	bFGF	All
7	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	14	14
9	NaN	NaN	NaN	NaN	NaN	NaN	6	NaN	NaN	6
10	NaN	NaN	14	NaN	NaN	NaN	NaN	NaN	NaN	14
13	NaN	NaN	4	NaN	NaN	NaN	NaN	NaN	NaN	4
15	NaN	NaN	7	NaN	NaN	NaN	NaN	NaN	NaN	7
16	NaN	NaN	NaN	NaN	NaN	7	NaN	NaN	NaN	7
21	NaN	NaN	NaN	NaN	10	NaN	NaN	NaN	NaN	10
22	NaN	NaN	7	NaN	NaN	NaN	NaN	NaN	NaN	7
23	NaN	NaN	NaN	21	NaN	NaN	NaN	NaN	NaN	21
24	NaN	NaN	NaN	NaN	NaN	NaN	3	NaN	NaN	3
26	NaN	NaN	NaN	NaN	NaN	NaN	16	NaN	NaN	16
33	10	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	10
34	NaN	NaN	7	NaN	NaN	NaN	NaN	NaN	NaN	7
37	NaN	NaN	NaN	NaN	NaN	NaN	5	NaN	NaN	5
41	NaN	NaN	1	NaN	NaN	NaN	NaN	NaN	NaN	1
53	NaN	NaN	7	NaN	NaN	NaN	NaN	NaN	NaN	7
54	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	7	7
55	NaN	NaN	NaN	NaN	NaN	NaN	5	NaN	NaN	5
62	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	17	17
63	NaN	NaN	NaN	NaN	NaN	NaN	17	NaN	NaN	17
64	NaN	10	NaN	NaN	NaN	NaN	NaN	NaN	NaN	10
69	NaN	NaN	NaN	NaN	NaN	NaN	NaN	7	NaN	7
79	NaN	NaN	NaN	NaN	NaN	NaN	NaN	7	NaN	7
86	NaN	NaN	NaN	3	NaN	NaN	NaN	NaN	NaN	3
87	NaN	NaN	5	NaN	NaN	NaN	NaN	NaN	NaN	5
88	NaN	NaN	5	NaN	NaN	NaN	NaN	NaN	NaN	5
89	NaN	NaN	NaN	NaN	NaN	NaN	14	NaN	NaN	14
97	NaN	NaN	NaN	NaN	NaN	NaN	12	NaN	NaN	12

# Fig 9-C1 (Continued: data graphing)

wdf_melted = wdf_24_a.melt(value_vars=['NGF', 'RA', 'bFGF', 'All'], 
                           var_name='Neurotrophic Factor', 
                           value_name='Days')

# Creating the horizontal boxplot
plt.figure(figsize=(10, 6))
sns.boxplot(x='Days', y='Neurotrophic Factor', data=wdf_melted, 
            order=['NGF', 'RA', 'bFGF', 'All'], palette='Set2')

plt.title('Days Incubated in Neurotrophic Factors', fontsize=18, fontweight='bold')
plt.xlabel('Days', fontsize=16)
plt.ylabel('Neurotrophic Factor', fontsize=16)
plt.yticks( fontsize=16)
plt.xticks( fontsize=16)
plt.grid(False) 

plt.savefig('neurotrophic_factors_boxplot.png', dpi=300, bbox_inches='tight')

plt.show()

# Fig 9-C2 ( data cleaning)

wdf_25 = df_1.iloc[1:,].loc[:,['Neurotrophic_factor','NF_concentration']].dropna(subset='NF_concentration')

# rearranging for graphing
wdf_25.reset_index(drop=True, inplace=True)
wdf_25_a = wdf_25.pivot(columns='Neurotrophic_factor', values='NF_concentration')
wdf_25_b = wdf_25_a[['RA', 'NGF', 'bFGF']].copy()
# dropping extreme outliers
wdf_25_b = wdf_25_b.drop([8,12,26,33])

# renaming column names

wdf_25_b.columns = wdf_25_b.columns.str.strip()

if 'RA' in wdf_25_b.columns:
    wdf_25_b['RA'] = wdf_25_b['RA'].str.replace('uM', '', regex=False).astype(float)
    wdf_25_b.rename(columns={
        'RA': 'RA (uM)',
        'NGF': 'NGF (ng/mL)',
        'bFGF': 'bFGF (ng/mL)'
    }, inplace=True)
else:
    print("Column 'RA' not found in the DataFrame.")

wdf_25_b.head(10)

Neurotrophic_factor	RA (uM)	NGF (ng/mL)	bFGF (ng/mL)
0	NaN	NaN	20
1	1.0	NaN	NaN
2	NaN	50	NaN
3	NaN	50	NaN
4	NaN	100	NaN
5	NaN	100	NaN
6	NaN	NaN	NaN
7	NaN	NaN	NaN
9	NaN	NaN	NaN
10	1.0	NaN	NaN

# Fig 9-C2 ( Continued : Visualization)

columns = ['RA (uM)', 'NGF (ng/mL)', 'bFGF (ng/mL)']

# Plotting each value with the corresponding number of occurrences
for column in columns:
    
    plt.figure(figsize=(8, 2))
    values = wdf_25_b[column].dropna()
    unique_values, counts = np.unique(values, return_counts=True)

    for val, count in zip(unique_values, counts):
        plt.scatter([val] * count, range(1, count + 1), alpha=0.7)
    plt.yticks([])  
    plt.xlabel(column)
    plt.title(f'Common Concentrations of {column}', fontsize=14, fontweight='bold')
    plt.gca().spines['top'].set_visible(False)
    plt.gca().spines['right'].set_visible(False)
    plt.gca().spines['left'].set_visible(False)
    plt.gca().spines['bottom'].set_visible(True)
    plt.grid(False)
    
    plt.show()

# Fig 9-D1 
wdf_26 = df.iloc[1:,].loc[:,['DC_Elec_field_mv/cm','Elec_stimu_hours','Elec_stimu_days']].dropna(subset='DC_Elec_field_mv/cm'	)
#dropping rows with insufficient data
wdf_26 = wdf_26.dropna(subset='DC_Elec_field_mv/cm').drop([9,35,28],axis= 0).reset_index().drop(columns= 'index')

# dataframe below was used to make the graph in Canva.com
wdf_26

	DC_Elec_field_mv/cm	Elec_stimu_hours	Elec_stimu_days
0	1000	14	14
1	100	28	7
2	2000	0.6	7
3	10	0.6	7
4	100	4	4
5	2000	7	7
6	100	2.3	7
7	100	2.5	5
8	200	1.5	3
9	100	2.3	7

# Fig 9-D2
wdf_27= df.iloc[1:,].loc[:,['Mag_field_mT','Mag_freq_Hz','Elec_stimu_hours','Elec_stimu_days']].dropna(subset='Mag_field_mT')
#dropping rows with insufficient data
wdf_27= wdf_27.drop([1,53,71],axis=0).reset_index().drop(columns= 'index')
# dataframe below was used to make the graph in Canva.com
wdf_27

	Mag_field_mT	Mag_freq_Hz	Elec_stimu_hours	Elec_stimu_days
0	15	100	1.25	5
1	1	50	24	7
2	3	2	1	3
3	7.7	Static	72	3
4	100	1	14	7
5	50	Static	336	14
6	200	Static	120	5
7	13	60	3.5	7
8	50	1053	10	10

# Fig 10-A (Pie chart core)

wdf_28 = df.iloc[1:,].loc[:,['Diff_asess_method']].dropna(subset='Diff_asess_method')

wdf_28_a = wdf_28['Diff_asess_method'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_28_b = pd.DataFrame(wdf_28_a, columns=['Diff_asess_method'])
wdf_28_b['Diff_asess_method'] = wdf_28_b['Diff_asess_method'].str.strip().str.title()

# counts below were used to make the core of the Pie chart
print(wdf_28_b[wdf_28_b.Diff_asess_method.str.contains('Protein Marker')].value_counts())
print('________________')
print(wdf_28_b[wdf_28_b.Diff_asess_method.str.contains('Morphology')].value_counts())
print('________________')
print(wdf_28_b[wdf_28_b.Diff_asess_method.str.contains('Gene Marker')].value_counts())
print('________________')
print(wdf_28_b[wdf_28_b.Diff_asess_method.str.contains('Electrophysiologica')].value_counts())

Diff_asess_method                       
Neural Protein Marker Analysis              68
In Vivo(Neural Protein Marker Analysis)      8
In Vivo (Neural Protein Marker Analysis)     6
In Vivo(Neural Protein Marker Analysis )     3
In Vivo(Neural Protein Marker Analysis       2
Name: count, dtype: int64
________________
Diff_asess_method
Morphology           40
Name: count, dtype: int64
________________
Diff_asess_method                    
Neural Gene Marker Analysis              37
Neural Gene Marker Analysis)              2
In Vivo (Neural Gene Marker Analysis)     1
Name: count, dtype: int64
________________
Diff_asess_method              
Electrophysiological Assessment    2
Name: count, dtype: int64

# Fig 10-A (Pie chart periphery)
wdf_29_a = df.iloc[1:,].loc[:,['Diff_morph']].dropna()
wdf_29_b = df.iloc[1:,].loc[:,['Genetic_marker_method']].dropna()
wdf_29_c = df.iloc[1:,].loc[:,['Protein_marker_method']].dropna()

# Counts obtained from below were used to make the periphery of the pie chart
print(wdf_29_a.value_counts())
print('________________')
print(wdf_29_b.value_counts())
print('________________')
print(wdf_29_c.value_counts())

Diff_morph                                           
Neurite Growth/Length                                    35
Other (myelination), Neurite Growth/Length                3
Other (myelination)                                       1
Synapsis formation/connectivity,Neurite Growth/Length     1
`                                                         1
Name: count, dtype: int64
________________
Genetic_marker_method
PCR                      38
Name: count, dtype: int64
________________
Protein_marker_method       
Immunostaining                  53
Immunostaining, Western blot     4
Western blot                     4
ELISA                            3
Immunostaining,CLSM              1
Immunostaining,western blot      1
In vivo (Immunostaining)         1
Name: count, dtype: int64

# Fig 10-B (Upper graph)
wdf_31 = df.iloc[1:,].loc[:,['genetic_marker_group']].dropna()
wdf_30_a = wdf_30['genetic_marker_group'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_30_b = pd.DataFrame(wdf_30_a, columns=['genetic_marker_group'])
wdf_30_b['genetic_marker_group'] = wdf_30_b['genetic_marker_group'].str.strip().str.title()

# the dataframe below was defined in R for graphing
wdf_30_b.value_counts() 

genetic_marker_group       
Neuronal Cells And Function    36
Astrocytes                     25
Nerual Stem Cells              18
Oligodendrocytes                8
Neurotrophic Factors            4
Name: count, dtype: int64

# Fig 10-B (lower graph)
wdf_31 = df.iloc[1:,].loc[:,['Protein_marker_group']].dropna()
wdf_31_a = wdf_31['Protein_marker_group'].str.split(',', expand=True).stack().reset_index(drop=True)
wdf_31_b = pd.DataFrame(wdf_31_a, columns=['Protein_marker_group'])
wdf_31_b['Protein_marker_group'] = wdf_31_b['Protein_marker_group'].str.strip().str.title()

# the dataframe below was defined in R for graphing
wdf_31_b.value_counts() 

Protein_marker_group       
Neuronal Cells And Function    63
Astrocytes                     28
Neural Stem Cells               8
Oligodendrocytes                8
Nerual Stem Cells               6
Neurotrophic Factors            2
Astrocyte                       1
Neurotrophic Factor             1
Name: count, dtype: int64

# Fig 10-C1 

# The cleaned data from "Clean" sheet is imported 
wdf_32 = df_1.iloc[1:,1:6]

# Exporting for graphing in R
wdf_32.to_csv('wdf_32.csv')

wdf_32.sample(10)

	Neural_Stem_Cells	Neuronal_Cells_and_Function	Oligodendrocytes	Astrocytes	Neurotrophic_Factors
32	NaN	MAP2	NaN	NaN	NaN
35	NaN	NaN	NaN	NaN	NaN
27	Nestin	Tuj1	NaN	NaN	NaN
82	SOX2	NaN	NaN	NaN	NaN
60	NaN	NaN	NaN	NaN	NaN
88	NaN	NaN	NaN	NaN	NaN
64	NaN	Other	NaN	NaN	NaN
17	Nestin	MAP2	NaN	NaN	NaN
100	NaN	NaN	NaN	NaN	NaN
21	NaN	NaN	NaN	NaN	NaN

# Fig 10-C2 

# The cleaned data from "Clean" sheet is imported 
wdf_33 = df_1.iloc[1:,7:12]
# Exporting for graphing 
wdf_33.to_csv('wdf_33.csv')

wdf_33

	Neural_Stem_Cells_P	Neuronal_Cells_and_Function_P	Oligodendrocytes_P	Astrocytes_P	Neurotrophic_Factors_P
1	NaN	Tuj1	NaN	NaN	NaN
2	Nestin	Tuj1	MBP	GFAP	NaN
3	NaN	Tuj1	NaN	NaN	NaN
4	Sox2	Tuj1	NaN	NaN	NaN
5	Sox2	MAP2	NaN	GFAP	NaN
...	...	...	...	...	...
96	NaN	Tuj1	NaN	NaN	NaN
97	NaN	NF	NaN	NaN	NaN
98	NaN	Other	NaN	NaN	NaN
99	NaN	NeuN	NaN	NaN	NaN
100	NaN	PSD95	NaN	NaN	NaN

100 rows × 5 columns

# Fig 11-A

wdf_34 =df_1.iloc[1:,].loc[:,['in_vivo_protein_mark']].dropna()

# Splitting the values in the column by commas and expanding into separate rows
wdf_34_a = wdf_34['in_vivo_protein_mark'].str.split(',', expand=True)
wdf_34_b = wdf_34_a.melt(value_name='Protein Markers').dropna()
wdf_34_b['Protein Markers'] = wdf_34_b['Protein Markers'].str.strip().str.upper().reset_index().drop(columns='index')
wdf_34_b = wdf_34_b.drop(columns='variable').dropna()
wdf_34_b.columns= ['Protein_Markers']
wdf_34_c = pd.DataFrame(wdf_34_b.Protein_Markers.value_counts()[:16]).reset_index()

wdf_34_c = wdf_34_c[wdf_34_c.Protein_Markers != 'IMMUNE/INFLAMMATION MARKERS']

#Exporting for graphing in R
wdf_34_c.to_csv('wdf_34_c.csv')

# Fig 11-B (core of the pie chart)

wdf_35= df.iloc[1:,].loc[:,['In_vivo_asess_method']].dropna()
wdf_35_a = wdf_35['In_vivo_asess_method'].str.split(',', expand=True)
wdf_35_b = wdf_35_a.melt(value_name='In_vivo_asess_method').dropna()
wdf_35_b['In_vivo_asess_method'] = wdf_35_b['In_vivo_asess_method'].str.strip().str.upper().reset_index().drop(columns='index')
wdf_35_b['In_vivo_asess_method'] = wdf_35_b.In_vivo_asess_method[wdf_35_b.In_vivo_asess_method != ""]
wdf_35_b.dropna().In_vivo_asess_method.value_counts() /len(wdf_35_b.dropna()) * 100

In_vivo_asess_method
HISTOLOGICAL ANALYSIS                35.625
PROTEIN MARKER ANALYSIS              30.000
BEHAVIORAL TESTING                   25.625
ELECTROPHYSIOLOGICAL MEASUREMENTS     8.125
GENE MARKER ANALYSIS                  0.625
Name: count, dtype: float64

# Fig 11-B (Continued: periphery of the chart)

wdf_36 = df.iloc[1:,].loc[:,['in_vivo_behavioral']].dropna()
wdf_36_a = wdf_36['in_vivo_behavioral'].str.split(',', expand=True)
wdf_36_b = wdf_36_a.melt(value_name='in_vivo_behavioral').dropna()
wdf_36_b['in_vivo_behavioral'] = wdf_36_b['in_vivo_behavioral'].str.strip().str.upper().reset_index().drop(columns='index')
wdf_36_b = wdf_36_b.dropna() 

print('Motor function periphery values')
print('__________________________________')
print('Motor function percentage')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('MOTOR', na=False)]) / len(wdf_36_b))
print('Motor function BBB')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('BBB', na=False)]) / len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('MOTOR FUNCTION', na=False)]))
print('Motor function SFI')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('SFI', na=False)]) / len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('MOTOR FUNCTION', na=False)]))
print('Motor function Walking track')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('WALKING TRACK', na=False)]) / len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('MOTOR FUNCTION', na=False)]))
print('Motor function fooprint')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('FOOT', na=False)]) / len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('MOTOR FUNCTION', na=False)]))
print('Sensory function percentage')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('SENSORY', na=False)]) / len(wdf_36_b))
print('Cognative/Emotional function percentage')
print(len(wdf_36_b[wdf_36_b['in_vivo_behavioral'].str.contains('COGNITIVE|OTHER', na=False)]) / len(wdf_36_b))

wdf_37 = df_1.iloc[1:,].loc[:,['elec_methods']].dropna()
wdf_37.value_counts()/ len(wdf_37)

Motor function periphery values
__________________________________
Motor function percentage
0.8913043478260869
Motor function BBB
0.4634146341463415
Motor function SFI
0.14634146341463414
Motor function Walking track
0.2682926829268293
Motor function fooprint
0.12195121951219512
Sensory function percentage
0.06521739130434782
Cognative/Emotional function percentage
0.043478260869565216

elec_methods                      
CMAP                                  0.315789
MEPs                                  0.157895
Motor and sensory-evoked potential    0.105263
CAP                                   0.052632
ICP                                   0.052632
LFP                                   0.052632
MAP                                   0.052632
NCV                                   0.052632
SCEP                                  0.052632
SEP                                   0.052632
Single-Unit recording                 0.052632
Name: count, dtype: float64